Artificial Intelligence, Expert Systems For Environmental And Energy Applications: The Study of Waste Products for Further Re-Cycling Into Finished Products

periodicdollsΤεχνίτη Νοημοσύνη και Ρομποτική

17 Ιουλ 2012 (πριν από 5 χρόνια και 1 μήνα)

360 εμφανίσεις


Environmental Informatics Archives, Volume 2 (2004), 722-730
EIA04-072
ISEIS Publication #002
© 2004 ISEIS - International Society for Environmental Information Sciences







Artificial Intelligence-Expert Systems for Environmental and Energy
Applications: The Study of Waste Products for Further Re-Cycling Into Finished
Products

A.A. Adegoke
1

*
and O. Okunowo
2

1.Channelsoft Model Consult, P.O.Box 3572,Mapo, Ibadan, Nigeria
2.Department of Computer Science, University of Ibadan, Nigeria.


Abstract. An expert systems known as automated waste content evaluator systems (AWCES) has been developed
to assist the environmental waste managers in determine the waste content materials like disposable nylon in
which water is packed (popularly called pure water in Nigeria), saw dust from planks, disposable plastics, used
straw for drinks, used paper est. needed for further re-cycling into finished products.
Given the waste material listings and their compositions, the expert systems produces a preliminary process flow
diagram showing all major waste content listings collected in an environment with their chemical combination and
life-span.
Chemical compositions of the waste products are tested against the standardized value via an interface to a
conventional process simulator. All of the quantities of the waste materials were calculated and the amount of
chemical needed for additions are estimated to complete the test for the recycling process. Various reports,
calculations and waste products datasheets are generated to summarize the result.
The system is flexible and allow the user to easily input all the parameters and change it, waste materials quantity
chemical combination calculation efficiently, automatic generation of reports and to produce outputs containing
whatever information is desired.
AWCES uses State Environmental Protection Agency for the research case study and make extensive use of
Prolog that runs on IBM compatible for evaluation and output generation before transfer to industry for processing.



*
Corresponding author:
kunle_adegoke@yahoo.com

E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
Keywords: Expert Systems, Artificial Intelligence, Environment, Nylon, Simulation, Recycling EST.


1. Introduction

Environmental sanitation has been introduced into Nigerian monthly activities as far back as early eighties
(1980’s), which has led to the cleaning of townships in Nigeria as a whole. Automated Waste Content Evaluator
System (AWCES) is an expert system that was suggested during the 2003 World Environmental Day by Oyo State
Environmental Protection Agency, Ibadan to assist the environmental waste managers of the agency for better
efficiency and maximizing waste products contents that can be used for further recycling to produce finished
products in the state.
Generally, expert system technology derived from the research discipline of Artificial Intelligence (AI) began in
Nigeria in the 1980’s.It is a branch of computer science capable of emulating human cognitive skills with a view
to solving problems, visual perception and language understanding. In view of these discoveries, computer, which
is now approaching the status of intelligent beings, is analyzed in this research work to help in simulating the
knowledge expertise of environmental waste managers at least in the areas of decision support in determine waste
materials that can be used for further recycling in the state.
The lassitude of computer has been seen in its in-ability to simulate human intelligence. Moreso, that the
potential capability and utilization of computer can only be most beneficial to mankind in its ability to do what can
be considered to be intelligence if done by human beings. Today’s modern technology in computing is now
heralding towards the fifth generation where the above-mentioned enervation is meant to be outdated.
With these, the application of expert system has been found in diverse range of organic chemistry, mineral
exploration and internal medicine but to mention a few. AWCES using expert system approach is developed to
simulate the decision-making process of environmental waste managers in determining whether waste contents
materials like disposable nylon, paper, used straw for drinks est. can further be recycled into finished products.
With this advent, Artificial Intelligence, Expert Systems, Neural Networks and Robotics have become the
computer terms to express the defacto that “computer is able to possess the innate characteristics of human
intelligence”. This new research innovation of computing as mentioned by the Japanese fifth generation of
computer has prompted the primary objective of the research to delve substantially into the study and design of an
Automated Waste Content Evaluator System (AWCES) to assist the environmental waste managers in decision-
making of waste content materials (solid) whether or not it can be recycled further into finished products.
AWCES help in promoting the economic development of the country, as there would be no need of importation
of these finished products. It will also help the environmental protection agency around the world in having a tidy
environment destitute of waste content littering.
Lastly, if anyone can’t escape the influence of computers in this modern-day technology, then introducing a
computer-based expert system to handle decision-making process for environmental protection will pave way for
global computerization that is moving round the world and total re-usable of our solid waste products.

2.Research Methodologies


723
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
The methods of data collection used for the expert system are personal interview and journals cum texts.
Environmental waste managers were contacted through personal interview for waste content material listings. This
same approach was also adopted in manufacturing company using these contents for details and percentage of
waste contents i.e. usable, name and amount of chemical requirements needed for recycling processes. Journals
were also contacted for chemical composition of the waste contents and their chemical balance equation

3.Artificial Intelligence-Expert System Review

The revolution of computer started with Charles Babbage when a design of analytical engine was developed for
numerical computation. Since then, development of computer has covered several milestones with each generation
ushering in more reliability, capability, efficiency and speed with lesser sizes and prices. Because of this
continuing progress, computer technology is now heralding on advances in the way computer can be used, not on
the electronic refinements that characterized the previous generation but on refinement that will regard computer
as being an intelligent processors of knowledge (Brightman and Dimsdale, 1986). However, there have been a lot
of controversies and criticisms about computer being an intelligent machine from the theologians and religious
leaders expressing the views that-the quest of human being to gain knowledge constitute a transgression against
the law of God especially in some areas relating to simulation of intelligence and believe that it is only through
humanistic incantations or unspeakable transactions with the underworld that such a machine can vindicate its
existence (Luger and Stubblefield 1998).
These opinions are totally against the opinion of Shelly (a scientist who shows the extent to which scientific
advances such as the work of Darvin and discovery of electricity had convinced even the non scientist in paper
titled “the wonder in Prometheus) that work of nature were not divine secrets, but could be broken down and
understood systematically.
Also, the development of renaissance thought initiated by the evolution of a different and powerful way of
thinking about humanity in relation to the natural world gave scientist more hopes that gradually wipe-out the
thought of critics that empiricism cannot replace mysticism as a means of understanding nature such as human
intelligence as regards to modelling of an intelligent machine (Luger and Stubblefield, 1989). Lindsay (1988)
explaining Yazdani positive writing in support of the AI research in his paper titled “ as birds are to airplanes, so
my brains be to beer cans” drew an analogy between thinking objects and flying objects in correspondence with
humans and birds that
“Artificial flight has progressed, not in direct imitation to natural flight as regards to biochemical properties of
birds but in regards to the aerodynamic properties of birds which can be simulated and likens the same situation
to AI that it seems reasonable that AI will eventually come to have a similar sort of relationship to natural
intelligence. That is to say AI should expect neither to imitate nor displace human intelligence, but simply to
operate within the area defined by a common set of principles called aerodynamics of intelligence.”
This analogy helps to explain the paradox that machines can think, but they cannot really “think”. She therefore
expressed the defacto that A.I will exists at one end of a continuum of intelligence with human intelligence at the
other end separated by degrees of difference not kinds. The paper was summarized by accepting the fact that
intelligence can be seen in machine without resolving to the theologians beliefs and content that “birds” do fly but
not in the same way as airplanes as computer will also be able to think but not like human beings.

724
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
The branch of Artificial Intelligence used is this research work is expert system; the computer programs that are
derive from a branch of computer science research called Artificial Intelligence (AI). AI's scientific goal is to
understand intelligence by building computer programs that exhibit intelligent behavior. It is concerned with the
concepts and methods of symbolic inference, or reasoning, by a computer, and how the knowledge used to make
those inferences will be represented inside the machine
One major insight gained from early work in problem solving was the importance of domain-specific
knowledge. A doctor, for example, is not effective at diagnosing illness solely because she possesses some innate
general problem-solving skill; she is effective because she knows a lot about medicine. Similarly, a geologist is
effective at discovering mineral deposits because he is able to apply a good deal of theoretical and empirical
knowledge about geology to the problem at hand. Expert knowledge is a combination of a theoretical
understanding of the problem and a collection of heuristic problem-solving rules that experience has shown to be
effective in the domain. Obtaining this knowledge from a human expert and coding it into a form that a computer
may apply to similar problems construct expert systems.
This reliance on the knowledge of a human domain expert for the system's problem solving strategies is a major
feature of expert systems. Although some programs are written in which the designer is also the source of the
domain knowledge, it is far more typical to see such programs growing out of a collaboration between a domain
expert such as a doctor, chemist, geologist, or engineer and a separate artificial intelligence specialist. The domain
expert provides the necessary knowledge of the problem domain through a general discussion of her problem-
solving methods and by demonstrating those skills on a carefully chosen set of sample problems.
The AI specialist, or
knowledge engineer
, as expert systems designers are often known, is responsible for
implementing this knowledge in a program that is both effective and seemingly intelligent in its behavior. Once
such a program has been written, it is necessary to refine its expertise through a process of giving it example
problems to solve, letting the domain expert criticize its behavior, and making any required changes or
modifications to the program's knowledge. This process is repeated until the program has achieved the desired
level of performance.
One of the earliest systems to exploit domain-specific knowledge in problem solving was DENDRAL,
developed at Stanford in the late 1960s (Lindsay et al. 1980). DENDRAL was designed to infer the structure of
organic molecules from their chemical formulas and mass spectrographic information about the chemical bonds
present in the molecules. Because organic molecules tend to be very large, the number of possible structures for
these molecules tends to be huge. DENDRAL addresses the problem of this large search space by applying the
heuristic knowledge of expert chemists to the structure elucidation problem. DENDRAL's methods proved
remarkably effective, routinely finding the correct structure out of millions of possibilities after only a few trials.
The approach has proved so successful that descendants of the system are used in chemical and pharmaceutical
laboratories throughout the world.
Whereas DENDRAL was one of the first programs to effectively use domain-specific knowledge to achieve
expert level problem-solving performance, MYCIN established the methodology of contemporary expert systems
(Buchanan and Shortliff 1984). MYCIN uses expert medical knowledge to diagnose and prescribe treatment for
spinal meningitis and bacterial infections of the blood.
MYCIN, developed at Stanford in the mid-1970s, was one of the first programs to address the problems of
reasoning with uncertain or incomplete information. MYCIN provided clear and logical explanations of its

725
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
reasoning, used a control structure appropriate to the specific problem domain, and identified criteria to reliably
evaluate its performance. Many of the expert system development techniques currently in use were first developed
in the MYCIN project.
Other classic expert systems include the PROSPECTOR program for determining the probable location and
type of ore deposits based on geological information about a site (Duda et al. 1979 ), the INTERNIST program for
performing diagnosis in the area of internal medicine, the Dipmeter Advisor for interpreting the results of oil well
drilling logs (Smith and Baker 1983), and XCON for configuring VAX computers. XCON was developed in 1981,
and at one time, that software configured every VAX sold by Digital Equipment Corporation. Numerous other
expert systems are currently solving problems in areas such as medicine, education, business, energy, environment,
geography, earth sciences, design, and science (Waterman 1986, Durkin 1994).
It is interesting to note that most expert systems have been written for relatively specialized, expert level
domains. These domains are generally well studied and have clearly defined problem-solving strategies. Problems
that depend on a more loosely defined notion of "common sense" are much more difficult to solve by these means.
In spite of the promise of expert systems, it would be a mistake to overestimate the ability of this technology.
In summary, AI is the part of computer science concerned with designing intelligent computer-system that
exhibit the characteristics we associate with human intelligent behaviour such as understanding, learning,
reasoning and solving problems. A.I. systems involve such higher mental processes as perceptual learning
memory organization and judgment reasoning. The combine feature of AI as a branch of engineering (i.e. building
intelligent artifacts) and cognitive science (i.e. the study of human information processing has extended the
discipline to be related to a wide range of academic subject areas such as computer science, psychology,
philosophy, environment, lingustics and engineering. A.I. application areas include:

• Natural Language Understanding
• Game Playing
• Automated Reasoning and Theorem Proving
• Modelling Human Performance-Cognitive Modelling and Neural Networks
• Planning and Robotics
• Languages and Environments for AI
• Machine Learning
• Expert Systems
(Luger and Stubblefield, 1989).

Figure 1 depicts some AI application areas as stated by McAllister (1987).
However, the identification of the usefulness of AI application areas in energy and environment has prompted
environmental scientists to identify the potential of expert system as one of the areas of AI to be used in simulating
environmental waste managers behaviour in recycling process of the waste products into finished products.





726
PROLOG LISP
ROBOTICS
NATURAL
LANGUAGE
LANGUAGES
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0


























Figure 1


4.Expert System Design and Implementation Issues

The architecture of Expert system can be seen from two perspectives-software and hardware. This was divulged
by Edmunds (1988), expert system is a computer, which consist of a group of computer programs meaning there
are software and hardware components as known of the computer system. However, the approach considered in
this research work delve basically into the software approach which any present computer system can adopt and
implement in creating an illusion that computer can possess the innate characteristics of human intelligence.
Henceforth, software approach sees expert system as knowledge based program that provides “expert-quality”
solution to problems in a specific domain. Also the architecture of expert system has been foretold to be
knowledge-rich even, if the methods of drawing inferences is poor-expressing the defacto that “ the power of
human expert resides in its knowledge and not on the biochemical method of reasoning thereby making expert

727
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
system to obviate from focusing more attention almost exclusively on the development of hardware components
that will provide facilities for executing these complex inferences-methods as does in some AI application
areas.(Luger and Stubblefield,1989;Edmunds,1988;Jackson,1990).

5.AWCES Discussion and Result

AWCES as an expert system was designed to support the environmental waste managers in decision-making
process of further recycling of waste products (such as nylon, paper, used straw) into finished products. The
system developed using PROLOG programming language has the characteristics to simulate human reasoning
about waste content materials, perform qualitative and quantitative reasoning and calculation over these waste
content, determine their reusability or not, and explain in terms of reports justify solutions or recommendation to
the managers. The system is designed to exhibit high performance in terms of speed and reliability in order to be a
useful, efficient and durable tool for decision-making. Environmental waste managers before development of this
expert system were concerned about how to effectively handle waste content materials, which were in vast amount
by the rapidly increasing population of the human race. Actually, many of these waste contents have bridged the
channel of flow of societal drainage system thus causing flood disaster within the State.
With the development of AWCES, waste managers could now give a sign of relief to societal problems of
handling waste content and boost also there efficiency in discharge of services thus making them to be part of the
global trend in computerization.
The stages of AWCES are explained thus: The system is user’s friendly. Users interact with the system through
a user interface. All waste content material listing collected by waste managers are entered as input data through
the user interface. These data are stored in the case specific data section of the expert system. This section also
contains implicit data declaration and initialization.
Afterward, a list of all waste materials content entered as input and their respective chemical compositions are
retrieve from the specific data section by the inference engine and are display to the user through an interface.
The inference engine (or problem solving component) here is use to simulate reasoning with both data specific and
the expert knowledge in the knowledge base. Since nylon is our case study in this scope, other explanation
following will be tailored on nylon.
The next process stage of the expert system sends the chemical composition of nylon (Adipic acid – COOH
(CH
2
)
4
COOH and Hexamethylenediamine NH
2
(CH
2
)
6
NH
2
) into a simulator for simulation process. This process
performs a calculation of chemical compositions of nylon, passed into the simulator, compares the values obtained
in grams to a standardized value for recycling. The standardized value is part of the data implicitly built in the
expert system.






Star
t


728
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0





















Display the waste content
and their com
p
osition
Nylon waste content
cannot be used
Nylon (adipic acid +
hexamethylenediamine
Sawdust
Used straw
To simulato
r
Value
<= 49%
Test for value of nylon
chemical composition
Paper
Enter Waste content materials
collected
Waste content collected
> = 50%


Calculate quantity of each
nylon component in kg/tons




Determine amount of
chemicals needed for




Display report of waste
content materials, etc



Sto
p




Figure.2

A conditional statement designed in the system carries out the comparison of the two values. If the
percentage value obtainable from the nylon calculation is less than 50%, the nylon waste content are

729
E
nvir
o
n
m
ental I
n
form
atics
Ar
chives, Volu
m
e
2
(
2004)
,
722-
73
0
discarded and hence cannot be recycle further into finish product. If the percentage is greater than or equal to
50%, the nylon waste content is transfer to the next process stage. At this stage, a chemical calculation is
performed on each composition of nylon to generate the quantity in kg/tons derivable from each composition
for recycling. In performing the calculation, the compound formula of each chemical composition is use in
performing this calculation.
Before nylon can finally be recycled into finished product, other chemical compounds will be added by the
industry to complete the recycling process. The amount of chemical needed for this completion must be
calculated in another stage. This is crucial as too much quantities or less than require amount will ultimately
affect the finished products in terms of quality.
Finally, the expert system via the knowledge base produces reports of all waste products collected, their
chemical compositions. Reports of chemical compound added and their quantity are also generated. This
system is an ongoing research study that can allow another specification, input and changes of its logical
representation of program running for further improvement to any other environmental waste products. It can
also further be sponsored to handle industrial decision support applications in environmental management.


References:

Payne McArtur, Developing Expert systems, 3rd Edition:
Ivan Bratko, Prolog programming for Artificial Intelligence. Prentice Hall.
Luger G.F. Stubblefield W.A. (1989). Artificial Intelligence and the Design of Expert System: The
Benjamin/cummings. Publishing Company, Inc: Massachusetts.
McAllister J. (1987). Artificial Intelligence and Prolog on Micro Computers. Edward Arnold Publishers
Ltd., London
The World Book Encyclopedia N.) Volume 14: World Book Inc. A Scot Fetzer Company.
Jackson P. (1990). Introduction to Expert Systems, 3
rd
edition. Addison-Wesley Publishing Company
Massachusetts.
Waterman D. A. (1986) A guide to Expert systems
Addison – Wesley Publishing Company
Quillan, J. K. (1987) Application of Expert Systems.
Addison Wesley Publishing Company, Sydney.

Reports:

A detailed report analysis of products of Nigerian Polythene Limited, Ibadan, Oyo State (NIPPOL) Nigeria.
Detailed report analysis of products of Trinics Poly Investment, Ibadan, Oyo State, Nigeria.



730